Lab Course 2014-2015

1. Lab Course2014-2015 CHL5207Y / 5208Y

2. Time/Location • Thursday afternoon 2-4 pm • HS100

3. Attendance • Sign in sheet will be distributed each week

4. Course Personnel • Tamara Arenovich (Director) • Rosane Nisenbaum (Instructor) • Derek Stephens (Instructor) • Marianne Tam (Administration)

5. Main contact information • Tamara.Arenovich@utoronto.ca • Marianne.Tam@utoronto.ca

6. Two streams • There are two streams of students: • CHL5207Y is for M.Sc. candidates in Biostatistics • CHL5208Y is for PhD candidates in Biostatistics. • Both courses are mandatory in their respective programs.

7. Course format • The course format, introduced in 2006/2007, includes: • weekly 2 hour lecture • 4 hours per week practicum

8. Course goals • Introduce statistical design and analysis topics that should be useful in your career as a consulting biostatistician • Provide hands-on experience with design and analysis issues encountered by applied statisticians in the workforce. This includes the analysis of data. • Provide experience in communicating statistical concepts in everyday language

9. Lectures

10. Term 1

11. Term 2 topics • Survival analysis • Bayesian analysis • Statistical Genetics • Simulation • Propensity Scores in observational studies • Writing an effective analysis report • What employers look for • Setting up your own consulting business

12. Practicum

13. Practicum • Matching process procedure • We want to give you experience applying and competing for placements • We want to facilitate a win-win scenario

14. Practicum Supervisor Responsibilities • Provide a student physical space and computing facilities at the supervisor’s place of work • Provide a student supervision / mentorship totaling 4 hours per week for the full academic year (October through to April) • Provide a series of “small” projects or one “large” project that exposes the student to practical issues faced by the applied statistician. This should involve analysis of data. • Evaluation of the student using a set template mid way through the year and at the end of the year.

15. Practicum Supervisors Confirmed

16. Practicum evaluation criteria

17. Presentations • 10 minutes in Term 1 • 15 minutes in Term 2 • Evaluation criteria to be announced later.

18. Course Grading

19. Report of the ASA Workgroup on Master’s Degrees 06 November 2012

20. Bob Rodriguez, while serving as President-elect of the American Statistical Association, formed a working group to address the following charge: “Develop guidelines, framed as learning outcomes, for master’s degree programs in statistics and biostatistics that are responsive to the needs of stakeholders who employ such graduates…These guidelines will assist master’s degree programs in statistics and biostatistics to align their curricula with desired outcomes.”

21. Executive Summary A phone interview of 29 recent graduates of Master’s programs in statistics and biostatistics and an email questionnaire of 19 employers of such graduates were conducted in 2012. The following seven recommendations emerged from a review of the responses to these surveys.

22. Recommendations • Recommendation 1: Graduates should have a solid foundation in statistical theory and methods. • Recommendation 2: Programming skills are critical and should be infused throughout the graduate student experience. • Recommendation 3: Communication skills are critical and should be developed and practiced throughout graduate programs.

23. Recommendations • Recommendation 4: Collaboration, teamwork, and leadership development should be part of graduate education. • Recommendation 5: Students should encounter non-routine, real problems throughout their graduate education. • Recommendation 6: Internships, co-ops or other significant immersive work experiences should be integrated into graduate education.

24. Recommendations • Recommendation 7: Programs should be encouraged to periodically survey recent graduates and employers of their recent graduates as a means of evaluating the success of their programs and to examine if other programmatic changes are warranted.

25. Knowledge/skills learned in graduate school that helped get first job • The responses to this question divided into two main groups. The largest response categories were: 1) programming; 2) general statistical/mathematical background; and 3) specific statistical tools/methods. These three categories were mentioned by 10-12 of the 29 recent graduates.

26. Knowledge/skills learned in graduate school that helped get first job • The second group included three different categories: 1) communication skills; 2) other statistical experience; and 3) graduate/teaching assistantship experience. These categories were mentioned by 3 of the 29 recent graduates. • The modal response category to the follow-up question about knowledge / skills that helped get a job after the first job was programming

27. Knowledge/skills that helped you perform your first job: • Broadly speaking, the respondents focused on three general areas: • Programming/Computing tools - The majority of respondents specifically mentioned SAS as important in performing their first job. Five respondents mentioned R, two mentioned SQL, and five mentioned programming skills generally.

28. Knowledge/skills that helped you perform your first job: • Statistics - The majority of the respondents mentioned something about having skills across a range of statistical models and methods. Specific areas mentioned more than once included linear and logistic models, data mining, Bayesian methods, time series, and survival analysis. GEE and mixed models were also mentioned. A couple of respondents discussed data preparation and data cleaning.

29. Knowledge/skills that helped you perform your first job: • Real-world skills - Most respondents discussed the importance of communication skills such as project reporting and presentation and dealing with clients, especially in the face of competing demands and deadlines. Several mentioned the usefulness of a statistical consulting class in this regard. One respondent also mentioned teaching skills.

30. Knowledge or skills graduates wished they had more of • Additional programming skills was the modal category (9 responses) to this question with specific statistical methods background (7 responses) and experience with ‘real data’/’big data’ (5 responses) following. Here, ‘programming’ referred to knowing other tools better (e.g. SQL, Excel) or deeper knowledge of a particular tool (e.g. ‘knew SAS better’). Other responses included people and consulting skills including communication/presentation skills (3 responses), project and time management

31. Characteristics of the top 2 candidates when interviewing 10 • Communicationwas mentioned by 14 of 19 employers with programming and statistical background mentioned by 11 and 10 employers, respectively. Communication reflected proficiency in written and oral reporting. Programming topics most commonly mentioned by respondents were SAS skills; but database skills and skills with other statistical programming environments, e.g. R, were mentioned by a few respondents. Statistical methods may have been a prerequisite before interviews were conducted.

32. Characteristics of the top 2 candidates when interviewing 10 • Attitude, thinking/problem solving and teamwork/collaboration/leadership followed in terms of frequency of being mentioned. Attitude was a category that included enthusiasm, interest, passion, professionalism and attention to detail. Thinking/problem solving addressed critical thinking skills applied to conceptualize and implement analyses in a logical flow.